- Email: [email protected]
Randomized clinical trial of surodex steroid drug delivery system for cataract surgery
Randomized Clinical Trial of Surodex Steroid Drug Delivery System for Cataract Surgery Anterior versus Posterior Placement of Two Surodex in the Eye Donald T. H. Tan, FRCS, FRCOphth1,2, Soon-Phaik Chee, FRCS, FRCOphth,1 Li Lim, MMed, FRCS,1 Julian Theng, MMed, FRCS,1 Manon Van Ede, MA1 Objective: To evaluate safety and antiinflammatory efficacy of placing two Surodex (Oculex Pharmaceuticals, Inc., Sunnyvale, CA) in the eye after cataract surgery in comparison with steroid eyedrops and to compare anterior versus posterior chamber placement. Design: Randomized, masked, controlled trial. Participants: One hundred four eyes of 104 Asian patients undergoing extracapsular cataract extraction with intraocular lens implantation were examined. Of these, 33 eyes of 33 patients served as control eyes (group A). Intervention: Two Surodex pellets were inserted in the anterior chamber (AC) of 35 eyes (group B), and two Surodex pellets were inserted in the ciliary sulcus of 36 eyes (group C) at the conclusion of surgery. Control eyes received neither Surodex nor a placebo implant, but were prescribed conventional 0.1% dexamethasone eyedrops four times daily for 4 weeks. Main Outcome Measures: Anterior chamber flare and cells were graded clinically at the slit lamp. Anterior chamber flare was assessed objectively with the Kowa FC500 Laser Flare Meter (Kowa Co. Ltd, Tokyo, Japan). Intraocular pressure and corneal endothelial specular microscopy with morphometric cell analysis were performed for up to 1 year after surgery. Results: Lower flare meter readings occurred in both Surodex groups at all postoperative visits, as compared with the dexamethasone eyedrop group, with statistical significance at days 4 (P ⫽ 0.001), 8 (P ⫽ 0.001), and 15 (P ⫽ 0.02). No difference in flare occurred between AC and ciliary sulcus placement. Clinical slit-lamp assessment of anterior chamber flare and cells showed no difference between Surodex-treated eyes and dexamethasone-treated eyes. Nine of 33 eyes (27.3%) in group A required steroid augmentation, as opposed to 4 of 71 eyes (5.6%) in groups B and C. Inflammatory symptoms were reduced in the Surodex-treated eyes, with statistical significance for ocular discomfort (P ⫽ 0.001), photophobia (P ⫽ 0.04), and lacrimation (P ⫽ 0.01). No complications occurred with Surodex-treated eyes, and no significant difference in endothelial cell loss was noted between Surodex-treated eyes and dexamethasone-treated eyes up to 1 year after surgery. Conclusions: Intraocular placement of two Surodex is a safe and effective treatment method to reduce intraocular inflammation after cataract surgery and clearly is superior to eyedrops in reducing inflammatory symptoms and aqueous flare as measured with the laser flare meter. No difference in efficacy between AC placement and ciliary sulcus placement of Surodex was detected in this study. Ophthalmology 2001;108: 2172–2181 © 2001 by the American Academy of Ophthalmology. The Oculex Drug Delivery System (DDS; Oculex Pharmaceuticals, Inc, Sunnyvale, CA) is a biodegradable system that provides sustained drug release after it is inserted into the eye at the conclusion of surgery. The DDS consists of a drug plus a biodegradable lactic acid and glycolic acid copolymer (poly[lactic-glycolic]-acid, or PLGA) that has been used as extensively as a synthetic absorbable suture Originally received: November 28, 2000. Accepted: July 12, 2001. Manuscript no. 200817. 1 Singapore National Eye Centre, Singapore Eye Research Institute, Singapore. 2 Department of Ophthalmology, National University of Singapore, Singapore. Presented in part as a poster at the annual meeting of the American Academy of Ophthalmology, San Francisco, California, October 1997.
2172
© 2001 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
material in general and ocular surgery as Vicryl (Johnson & Johnson Medical Inc., Jurong, Singapore) has been used, and the polymer has been shown to degrade safely by hydrolysis into the natural byproducts of lactic acid and glycolic acid, which are metabolized.1 The potential advantages of DDS over conventional postoperative eyedrops are
Supported by the Singapore National Medical Research Council (grant no.: STB/0050/1991), Singapore, and Oculex Pharmaceuticals, Inc., Sunnyvale, California. The authors do not have a financial interest in this or competing products. Correspondence to Donald T. H. Tan, FRCS, FRCOphth, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 0316. E-mail: snecdt@ pacific.net.sg ISSN 0161-6420/01/$–see front matter PII S0161-6420(01)00839-9
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery (1) significantly lower drug concentrations required (and hence reduced systemic side effects and systemic drug toxicity), (2) controlled anterior chamber (AC) drug concentrations with near-zero order pharmacokinetics, and (3) reduced complications related to patient noncompliance with eyedrop administration. Surodex is a DDS formulation containing 60 g of dexamethasone designed to reduce postoperative inflammation after cataract surgery. Surodex has previously been shown to be both safe and effective in delivering dexamethasone into animal as well as human eyes. Kochinke et al [Invest Ophthalmol Vis Sci 35(Suppl):4474, 1994] showed that Surodex provided effective controlled drug delivery in the AC with near-zero order pharmacokinetics over a 7-day period in rabbit eyes. Wong et al also demonstrated the safety of Surodex in the AC in rabbits in a Food and Drug Administration phase I toxicology study [FDA study protocol no. 94001], and Surodex was initially implanted in six human eyes undergoing phacoemulsification in a Food and Drug Administration phase I pilot human eye trial [FDA study protocol no. C95-001]. We previously performed a randomized, partially controlled human clinical trial comparing Surodex with conventional 0.1% dexamethasone eyedrops in cataract patients undergoing extracapsular cataract extraction (ECCE) with intraocular lens (IOL) implantation.2 Results showed that placement of one Surodex in the inferior angle of the AC was significantly more effective in reducing aqueous flare levels after surgery, as measured with laser flare photometry for up to 30 days after surgery. In addition, Surodex was found to be safe, with no complications occurring within the trial, no local reaction around the site of Surodex placement, and no significant difference in corneal endothelial cell counts between Surodex and eyedrop treatment arms 1 year after surgery. The aim of this clinical trial was to evaluate the safety and antiinflammatory efficacy of two DDS pellets of Surodex in the eye, as compared with conventional steroid eyedrops, and to evaluate the relative safety and efficacy of Surodex placement in the AC versus the posterior chamber (PC).
Materials and Methods One hundred four patients undergoing elective ECCE with IOL implantation were enrolled in a prospective, double-masked, randomized, partially controlled clinical trial during the period of August 1996 through January 1998 at the Singapore National Eye Centre, the tertiary referral national ophthalmic institute in Singapore. The clinical trial protocol and informed consent form were approved by the Singapore National Eye Centre Ethics and Research Committees, and written informed consent was obtained. Study conduct conformed to scientific principles embodied in the World Medical Association Declaration of Helsinki, as revised in 1989. Inclusion criteria included consecutive patients with uncomplicated, senile cataract listed for ECCE and IOL surgery by the two study consultant surgeons (D.T.H.T., S.P.C.). Exclusion criteria included patients taking oral or topical antiinflammatory agents, with diabetes mellitus, with a history of steroid-induced ocular hypertension, with hypermature cataracts, and eyes with uveitis, glaucoma, and corneal disease. Only one eye of each patient was enrolled in the trial.
Treatment Assignation and Study Masking Study participants were stratified into three categories according to age (50 –59 years, 60 – 69 years, and 70 –79 years). Eyes within each category were assigned randomly to receive one of three postoperative steroid treatment regimens from a table of computergenerated random numbers. Group A patients received conventional 0.1% dexamethasone eyedrops four times daily for a period of 30 days after surgery. Group B patients received two Surodex pellets inserted into the inferior angle of the AC at the conclusion of surgery and were prescribed normal saline placebo eyedrops. Group C patients received two Surodex pellets inserted into the inferior ciliary sulcus of the PC at the conclusion of surgery and were prescribed normal saline placebo eyedrops. Patients and study investigators were masked as to treatment regimens and, to reduce surgical bias, treatment groups were assigned by a study nurse coordinator who was the only person during the trial who was aware of patients’ treatment group status. The study code was kept in a locked drawer by the study nurse coordinator in the center’s clinical trials office. Surgeons were only informed of individual patients’ group status immediately before wound closure, at the conclusion of surgery, when Surodex insertion was required. After surgery, to reduce observer bias, all clinical assessment of intraocular inflammation was performed by a separate, masked investigator or ophthalmic technician who was unaware of patients’ group status. No inadvertent unmasking of patients occurred during the study.
Statistical Analysis and Sample Size Calculation The primary endpoint to compare relative efficacy of two Surodex pellets either in the AC (group B) or PC (group C) versus steroid drops was defined as the degree of AC flare, as measured objectively by the laser flare meter at the preset time points of 4, 8, 15, 30, and 90 days after surgery. A secondary outcome measure was chosen to be the semiquantitative measurement of cells and flare by slit-lamp grading. Treatment comparisons were performed with two-tailed significance levels set at P ⬍ 0.025 (Bonferroni-Holm) for all pair-wise comparisons, i.e., combined groups B and C versus group A, group B versus group A, and group C versus group A. The degree of inflammation is a continuous variable, and the Wilcoxon signed-rank tests of significance for expected skewed distribution of data were used for analysis of results. Both parametric and nonparametric tests were performed and showed consistent results. A secondary analysis of the data was performed with repeated measures analysis of variance with borderline covariate adjustment. This was performed for analysis of laser flare meter values and for endothelial cell counts. Study results were also compared directly with those obtained in our previous study with one Surodex insertion, which had a similar study methodology. All results were analyzed initially on an intention-to-treat basis. However, as the study progressed, it became clear that significant pellet migration to the AC was noted when Surodex was placed in the ciliary sulcus of the PC (group C), and gonioscopy at day 8 revealed that of the 36 eyes in group C, 15 eyes (58%) had evidence of either one or two Surodex present in the AC. With this in mind, a reclassification of group C was performed based on gonioscopic status at day 8: group C1 denoted eyes with evidence of one Surodex in the AC (and therefore one additional Surodex in the PC), whereas those group C eyes with no Surodex seen on day 8 gonioscopy (i.e., no pellet migration occurred) were categorized as group C2. Those eyes with two Surodex migrated to the AC were recategorized as group B. All study analyses conformed to this recategorization of treatment groups. Sample size calculations were based on available data on laser flare meter and slit-lamp inflammatory grading results accrued
2173
Ophthalmology Volume 108, Number 12, December 2001 from our previous trial using one Surodex. For laser flare meter values, to demonstrate statistical significance with a type I error (␣) of 0.025 and power of 80%, for an expected difference of 15 photons ms⫺1 (standard deviation, 15.0), the minimal sample size per treatment group (1:1:1 allocation) was 20 patients. For slitlamp grading, to demonstrate statistical significance with a type I error (␣) of 0.05 and power of 80%, for an expected difference of 1.0 (standard deviation, 1.5), the minimal sample size was 36 patients.
Definition of Clinical Therapeutic Failures Cases requiring additional use of antiinflammatory agents outside the study regimen at any postoperative period were considered as clinical therapeutic failures. Judgment was based on a subjective clinical evaluation of the degree of ocular inflammation present at either the ocular surface, limbal wound, or intraocularly. This clinical decision to augment steroid therapy was made by the study surgeons and was independent of the study’s postoperative investigator inflammatory grading.
Preoperative Assessment Before surgery, unaided and aided Snellen visual acuity, manifest refraction, Goldmann tonometry, and slit-lamp and fundus examinations were performed in all cases. Preoperative endothelial specular microscopy with endothelial cell morphometric analysis of the central and inferior corneal endothelium was performed with the Konan NONCON ROBO noncontact specular microscope (Konan, Osaka, Japan). Preoperative laser flare meter measurements were performed with the Kowa FC500 Laser Flare Meter (Kowa Co. Ltd., Tokyo, Japan).
Laser Flare Meter Measurement Technique All eyes were dilated with one drop of tropicamide 1% eyedrop before laser flare meter measurement, and all measurements were performed within 30 minutes of instillation. Laser flare measurement was performed in a totally dark room by a trained technician who was masked to patient group status, and a minimum of 10 recorded flare values were averaged for each visit.
Preoperative Medication All eyes were dilated before surgery with two drops of phenylephrine 10% eyedrop administered 5 minutes apart, three drops of tropicamide 1% eyedrop administered 3 minutes apart, and six drops of flurbiprofen sodium 0.03% eyedrop, administered 10 minutes apart. Preoperative antibiotic prophylaxis consisted of one drop of chloramphenicol 0.5% eyedrop administered 1 hour before surgery.
Surgical Procedure and Insertion of Surodex All cases received either retrobulbar or peribulbar anesthesia. Extracapsular cataract extraction was performed through a superior limbal incision, using a conventional can-opener capsulotomy technique, and the nucleus was delivered under Healon. Manual irrigation-aspiration was carried out with a Simcoe cannula, and insertion of a one-piece polymethyl methacrylate IOL (Cilco MZ60BD) was performed in all cases. After IOL insertion, the study envelope was opened by the operating theater circulating nurse to ascertain group status of the patient. For group B patients, two Surodex pellets were inserted into the AC with the use of Kelman-McPherson forceps and gently positioned in the inferior
2174
angle with the help of a Sinskey hook (Figs 1 and 2). For group C patients, two Surodex pellets were inserted into the ciliary sulcus of the PC with the use of Kelman-McPherson forceps. A Sinskey hook was used to push the pellets inferiorly out of sight beneath the iris. Wound closure was then effected with the use of 6 to 7 interrupted 10-0 nylon sutures, and Healon was carefully aspirated from the AC. A subconjunctival injection of gentamicin (4 mg) completed the procedure. All surgery was videotaped and archived.
Postoperative Medication All patients were prescribed chloramphenicol 0.5% eyedrop four times daily for 30 days, and oral acetazolamide 500 mg immediately after surgery and 250 mg twice daily for 48 hours after surgery. Patients were prescribed masked study eyedrops by the study nursing coordinator on the day of surgery and were instructed to apply four drops per day at equal intervals for 30 days. Study eyedrops for group A patients consisted of dexamethasone 0.1% eyedrop (Gutt Maxidex), whereas group B and C patients were prescribed 0.9% sodium chloride eyedrops (Xepa-Soul Pattinson (M) Sendirian Berhad) as placebo. Identical eyedrop bottles and labeling were used for all groups, and the surgeon, postoperative investigator, and patients were masked as to the study eyedrops.
Postoperative Evaluation Patients initially were requested to return for follow-up visits on days 1, 2, 4, 8, 15, 30, 60, and 90 after surgery. Subsequent to journal reviewers’ request for longer follow-up for our first Surodex study, it was decided that patients should also be followed up for at least 12 months, and the additional 12-month visit was added. Attempts also were made to recall all patients who had already been seen for their final 90-day visit. Examination at each postoperative visit included unaided and aided Snellen visual acuity, Goldmann applanation tonometry, and slit-lamp examination. Laser flare meter measurement was performed at each visit up to the day 90 visit. A separate, masked investigator performed slit-lamp grading of flare and cells. Gonioscopy was initially performed on the fourth postoperative day and at every subsequent postoperative visit (Figs 3, 4, 5, and 6). At the 90-day visit, fundus examination and manifest refraction were performed in all cases. Central and inferior corneal endothelial specular microscopy with endothelial cell morphometric analysis was performed after surgery at the 90-day visit and at the 1-year visit.
Clinical Assessment of Ocular Inflammation At each visit, ocular inflammation was assessed clinically by direct questioning for symptoms of inflammation (the presence of ocular discomfort or pain, tearing, and photophobia), as well as slit-lamp grading of intraocular cells and flare, conjunctival hyperemia, and ciliary injection according to established grading scales previously described by Kraff et al.3 Anterior chamber cell and flare grading scores were added together as a combined inflammatory score. Particular attention was paid to masking of the investigator grading inflammation, which was achievable because the technique of cell and flare assessment that involved the use of a minimum slit aperture of 0.2 mm and centration of the slit beam in the AC precluded a view of the inferior AC.
Results One hundred four eyes of 104 patients were enrolled in the study from August 1996 through January 1998, with 33 eyes in group A,
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery
Figure 1. Slit-lamp photograph of an isolated instance in which two Surodex are visible in the inferior angle. Note that in most such cases, Surodex pellets were sited deeper within the angle and were not visible on slit-lamp examination. Figure 2. Gonioscopic view of inferior angle shows two Surodex placed adjacent to each other in the inferior angle. Figure 3. Gonioscopic view of inferior angle showing Surodex remnants in early dissolution. Note that the peripheral cornea adjacent to Surodex remnants remains clear, and the inferior angle appears normal with no evidence of peripheral synechiae or scarring. Figure 4. Gonioscopic view of inferior angle showing Surodex remnants in late dissolution (6 months). Minute remnants or residue remains in the angle. Figure 5. Gonioscopic view showing normal inferior angle structures after complete Surodex dissolution. Figure 6. Slit-lamp photograph showing mild globular, lobulated appearance of dissolving Surodex. No adjacent corneal damage or clouding is seen.
2175
Ophthalmology Volume 108, Number 12, December 2001 Table 1. Mean Laser Flare Meter Baseline Values: Repeated Measurements Adjusted for Baseline Value
Group
P Value for Repeated Measurements
Post Hoc Tests*
Interaction† Time* Group
A vs. B vs. C1 vs. C2 A vs. B vs. C1 ⫹ C2 A vs. B ⫹ C
0.053 0.025 0.007
NS A vs. B, P ⫽ 0.027 —
NS 0.033 0.008
NS ⫽ not significant. *Bonferroni. † Greenhouse-Geiser.
35 eyes in group B, and 36 eyes in group C. The mean ages of patients in groups A, B, and C were 67.4 years, 66.4 years, and 68.3 years, respectively. Forty-six percent of patients were male and 54% were female. Most patients were Chinese (85%), whereas 7%, 4%, and 4% of patients were Malays, Indians, and other racial groups, respectively. This approximated the racial mix in Singapore. There were no statistically significant differences in age, race, or gender in the study groups. All 104 patients completed the initial study exit period at 90 days, but only 87 patients (83.7%) attended the 1-year recall visit. Eighty-five percent of group A patients, 82% of group B patients, 100% of group C1 patients and 80% of group C2 patients completed the additional 1-year visit. Comparison of visual acuity, IOP, symptoms, and endothelial cell counts obtained during the last study visit between the 17 patients who did not complete the month 12 follow-up and the other 87 patients revealed no statistically significant differences between the two groups.
Laser Flare Meter Results Mean laser flare meter values for individual study arms were analyzed at all time points up to 3 months after surgery. Preoperative flare values showed a difference between treatment groups, with generally higher preoperative flare values in the Surodex treated eyes. This was confirmed with repeated measurements, including baseline flare value as a covariant (Table 1). After surgery, mean flare values were lower in the Surodex groups (B and C) as compared with the eyedrops group (A) at all visits up to day 90 (Table 2; Fig 7). Flare reduction in the Surodex groups attained statistical significance at days 4, 8, and 15 (P ⫽ 0.001, P ⫽ 0.001, P ⫽ 0.02, respectively). When groups B and C were compared individually with group A, statistical significance was also achieved at these same time points. However, when groups C1 and C2 were compared with group A, no statistical significance was achieved in the group C1, whereas statistical significance was achieved in group C2 at days 4, 15, 30, and 60. Analysis of mean flare values between Surodex groups (B, C1, and C2) showed no statistically significant difference in efficacy between AC or PC placement of Surodex (Table 2).
Clinical Inflammation and Symptom Scores No significant differences in slit-lamp combined cell and flare scores occurred between the three treatment arms at any study time point (Table 3; Fig 8). No differences in ciliary injection or hyperemia were noted between treatment arms (data not shown). However, evaluation of inflammatory symptoms showed a clear difference between the eyedrop-treated group and Surodex groups for discomfort, photophobia, and lacrimation (Table 4). The percentage of patients who reported these symptoms at individual
2176
visits was significantly higher in the eyedrop-treated group compared with the Surodex-treated groups.
Dissolution of Surodex As in our previous study, we noted that the Surodex pellets gradually dissolved into ovoid or circular remnants at the site of original insertion and finally disappeared completely or left tiny residues (Figs 3, 4, and 5). In some instances, a lobulated, semitranslucent change occurred, associated with a slight increase in size, possibly suggesting pellet hydration (Fig 6). At the day 90 visit, gonioscopy confirmed that 57 eyes of the total of 71 eyes with at least one Surodex originally observed in the AC angle (80%) still had evidence of pellet remnants in the angle. At 6 months, pellet remnants were noted to be less, with 41 eyes (57%) still exhibiting remnants in the inferior angle, whereas at 1 year, 22 eyes (31%) had minute residual pellet remnants or residues.
Localized Angle Changes As in our previous study, mild localized angle changes were noted at or adjacent to pellet remnants in a percentage of eyes. In all cases, these were mild degrees of focal peripheral anterior synechiae with a notched appearance that were not progressive and did not result in pupil distortion. These lesions were visible only on careful gonioscopic examination, were attributed to focal adherence of iris to the DDS pellet, and were deemed clinically insignificant. Of the 71 eyes with Surodex present in the angle, 22 eyes (31%) exhibited these mild focal peripheral anterior synechiae lesions.
Corneal Endothelial Cell Morphometric Analysis No reduction in endothelial cell count occurred in Surodex-treated eyes, as compared with the dexamethasone eyedrop-treated eyes (Table 5). Because the baseline value was associated significantly with follow-up values, repeated-measures analysis of variance with borderline covariate adjustment was performed, with post hoc testing confirming a lack of any significant difference (Table 1). Similar analysis for inferior endothelial cell count also showed no significant differences between treatment arms (data not shown). In addition, the percentage change in 3-month and 1-year postoperative central and inferior endothelial cell count values from preoperative baseline values was also compared (data not shown). No significant differences in percentage change of cell counts were noted among all three treatment arms at both the 90-day visit and the 1-year visit.
Therapeutic Failures There were 9 eyes in group A (27%) that clinically required steroid augmentation because of excessive postoperative inflammation or
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery Table 2. Mean Laser Flare Meter Values: Comparison between Eyedrops and Surodex Groups and Position of Surodex Preoperative
Day 4
Day 8
Day 15
Day 30
Day 60
Day 90
Group A N ⫽ 33 N ⫽ 32 N ⫽ 32 Mean 6.15 33.99 24.52 SD 2.3 34.1 16.3 Median 5.6 24.3 18.0 Group B N ⫽ 49 N ⫽ 46 N ⫽ 46 Mean 6.27 20.19 16.68 SD 1.4 7.5 8.5 Median 6.2 18.2 14.3 Group B ⫹ C N ⫽ 71 N ⫽ 67 N ⫽ 65 Mean 6.35 20.75 17.79 SD 1.7 10.2 12.7 Median 6.1 18.2 14.2 Group C1 N⫽7 N⫽6 N⫽6 Mean 7.64 18.18 14.60 SD 3.2 3.7 4.9 Median 6.0 18.3 12.8 Group C2 N ⫽ 15 N ⫽ 15 N ⫽ 13 Mean 6.00 23.47 23.20 SD 1.4 17.0 23.2 Median 5.8 18.2 14.4 Group C1 ⫹ C2 N ⫽ 22 N ⫽ 21 N ⫽ 19 Mean 6.52 21.96 20.48 SD 2.2 14.6 19.6 Median 5.9 18.2 14.2 Significance levels for comparison between eyedrops and Surodex groups A vs. B ⫹ C 0.00 0.00 A vs. B 0.01 0.00 A vs. C1 ⫹ C2 0.01 0.04 A vs. C1 0.21 0.26 A vs. C2 0.04 0.21 Significance levels for comparison between anterior and posterior positions B vs. C2 0.86 0.65 B vs. C1 0.81 0.59 B vs. C1 ⫹ C2 0.80 0.94
N ⫽ 32 21.38 22.4 16.7 N ⫽ 49 13.47 5.5 12.7 N ⫽ 71 13.50 6.1 12.5 N⫽7 13.81 5.9 13.3 N ⫽ 15 13.33 8.3 11.1 N ⫽ 22 13.49 7.4 11.7
N ⫽ 32 12.78 6.1 11.6 N ⫽ 48 10.32 3.9 9.2 N ⫽ 70 10.14 4.0 9.0 N⫽7 10.31 3.1 10.0 N ⫽ 15 9.47 5.0 8.0 N ⫽ 22 9.74 4.4 8.1
N ⫽ 31 11.19 4.6 10.1 N ⫽ 48 9.69 3.7 9.2 N ⫽ 70 9.36 3.5 8.9 N⫽7 9.27 1.6 9.7 N ⫽ 15 8.25 3.1 7.3 N ⫽ 22 8.57 2.7 8.1
N ⫽ 33 9.09 3.5 8.8 N ⫽ 47 8.86 2.4 8.8 N ⫽ 69 8.70 2.3 8.7 N⫽7 8.71 1.7 8.5 N ⫽ 15 8.19 1.9 8.3 N ⫽ 22 8.36 1.9 8.4
0.02 0.04 0.04 0.34 0.04
0.07 0.15 0.05 0.46 0.04
0.05 0.16 0.03 0.53 0.02
0.88 0.76 0.86 0.85 0.72
0.43 0.96 0.56
0.16 0.85 0.31
0.09 0.67 0.26
0.31 0.96 0.41
SD ⫽ standard deviation. P values in bold type are statistically significant (P ⬍ 0.05).
significant remaining intraocular inflammation at 1 month, and therefore were clinical therapeutic failures. In contrast, only two eyes among the Surodex-treated eyes (one in group B, one in group C) were clinical therapeutic failures. This reduction in therapeutic failures in Surodex-treated eyes as compared with eyedrop-treated
eyes was statistically significant (P ⫽ 0.001, Pearson’s chi-square test). No sight-threatening complications occurred in these eyes with therapeutic failures.
Intraoperative Complications Inferior zonulysis with vitreous loss occurred in one group B patient. Anterior vitrectomy with AC IOL implanted successfully, and Surodex placements in the inferior chamber angle were performed. After surgery, no complications were encountered in this patient, and visual acuity attained was 6/6. No other intraoperative complications were encountered.
Postoperative Complications
Figure 7. Line graph showing mean laser flare meter values for all groups.
No severe complications were encountered after surgery. In group A, mild complications included transient raised IOP in two eyes (one requiring a short course of topical -blockers), blepharitis (n ⫽ 1), keratitis medicamentosa (n ⫽ 1), and fibrotic posterior capsule opacification in two eyes, with visual acuities of 6/12 and 6/18 (neodymium:yytrium–aluminum– garnet laser capsulotomy not required). One group A patient was noted to have excessive inflammation at the wound margin, together with a localized scleritis, which responded well to a 1-week course of oral prednisolone. In group B, mild complications included a mild surgical hyphema clot in one patient on day 1, which resolved, corneal
2177
Ophthalmology Volume 108, Number 12, December 2001 Table 3. Combined Cell and Flare Slit-lamp Grading Preoperative
Day 4
Day 8
Day 15
Cell grading (slit-lamp) Group A N ⫽ 33 N ⫽ 33 N ⫽ 33 N ⫽ 32 Mean 0 2.76 2.21 1.66 SD 0 1.0 0.8 0.7 Median 0 3 2 2 Group B ⫹ C N ⫽ 71 N ⫽ 71 N ⫽ 71 N ⫽ 71 Mean 0.01 3.13 2.24 1.63 SD 0.1 0.9 1.0 0.6 Median 0 3 2 2 Flare grading (slit-lamp) Group A N ⫽ 33 N ⫽ 33 N ⫽ 33 N ⫽ 32 Mean 0 1.27 0.97 0.53 SD 0 0.5 0.6 0.7 Median 0 1 1 0 Group B ⫹ C N ⫽ 71 N ⫽ 71 N ⫽ 71 N ⫽ 71 Mean 0.03 1.13 0.89 0.48 SD 0.2 0.6 0.6 0.6 Median 0 1 1 0 Combined cell and flare Group A N ⫽ 33 N ⫽ 33 N ⫽ 33 N ⫽ 32 Mean 0 4.03 3.18 2.19 SD 0 1.3 1.2 1.0 Median 0 4 3 2 Group B ⫹ C N ⫽ 71 N ⫽ 71 N ⫽ 71 N ⫽ 71 Mean 0.07 4.21 3.13 2.11 SD 0.3 1.3 1.4 1.0 Median 0 4 3 2 Significance levels for comparison between eyedrops and Surodex groups (A vs. B ⫹ C) Cell 0.06 0.98 0.85 Flare 0.22 0.51 0.74 Combined 0.26 0.97 0.70
Day 30
Day 60
Day 90
N ⫽ 32 0.91 0.8 1 N ⫽ 70 0.97 0.7 1
N ⫽ 31 0.84 0.7 1 N ⫽ 70 0.70 0.6 1
N ⫽ 32 0.50 1.0 0 N ⫽ 69 0.51 0.6 0
N ⫽ 32 0.19 0.4 0 N ⫽ 70 0.19 0.4 0
N ⫽ 31 0.13 0.3 0 N ⫽ 70 0.11 0.3 0
N ⫽ 32 0.06 0.3 0 N ⫽ 69 0.04 0.2 0
N ⫽ 32 1.09 1.0 1 N ⫽ 70 1.14 0.9 1
N ⫽ 31 0.97 0.8 1 N ⫽ 70 0.77 0.7 1
N ⫽ 32 0.36 0.6 0 N ⫽ 69 0.55 0.7 0
0.61 0.98 0.64
0.37 0.83 0.22
0.49 0.70 0.15
SD ⫽ standard deviation. Note: Wilcoxon rank sum test.
abrasion (n ⫽ 1), lower eyelid entropion (n ⫽ 1), and blepharitis (n ⫽ 1). In group C, mild complications included a mild surgical hyphema clot in one patient on day 1, which resolved, a peaked pupil resulting from an anterior capsule tag, raised IOP in one eye that resolved with a short course of topical -blockers, and blepharitis in two eyes. In addition, in two eyes, pellet migration into the
visual axis occurred between the posterior convexity of the IOL and the posterior capsule. However, pellet remnants dissolved without residual posterior capsule opacification, and no visual complaints or disturbances were noted by either patient on direct questioning. Peripheral anterior synechiae at the site of pellet implantation in the inferior chamber angle was noted in 18 of 71 eyes (25%) with Surodex in the AC angle. In all cases, the area of synechiae was minimal, involving less than one-half a clock hour of angle, and in all cases this was deemed not clinically significant. No progression of the degree of synechiae was noted, and no complications arose from these areas of synechiae.
Visual Acuity
Figure 8. Line graph showing mean combined cell and flare clinical slit-lamp results between treatment arms.
2178
At day 90, all eyes attained 6/12 or better corrected or uncorrected visual acuity, with the exception of one eye from group A that attained a best-corrected visual acuity of 6/18 because of the presence of mild fibrotic posterior capsule opacification. No loss of visual acuity attributable to Surodex insertion was seen in groups B or C. At the 1-year follow-up, two eyes (one in group B and one in group C) were noted to have foveal changes consistent with age-related macular degeneration, and attained best-corrected visual acuity of 6/18 and 6/15, respectively. In addition, one group A patient attained best-corrected visual acuity of 6/18, attributed to a combination of asteroid hyalosis and high corneal astigmatism of 5 diopters. No loss of acuity was related to Surodex insertion.
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery Table 4. Inflammation Symptoms: Percentage within Groups Day
Discomfort A B C1 ⫹ C2 Pain A B C1 ⫹ C2 Photophobia A B C1 ⫹ C2 Lacrimation A B C1 ⫹ C2 Others A B C1 ⫹ C2 Significance levels
1
4
8
15
30
60
90
% % %
24 4 5
27 20 9
21 10 5
25 12 9
19 4 9
16 6 0
3 0 0
% % %
3 0 5
3 0 0
0 0 0
6 2 0
3 0 0
0 2 0
0 0 0
% % %
6 0 5
3 0 0
3 0 5
3 0 0
0 0 0
0 0 0
0 0 0
% % %
21 8 9
24 27 23
33 31 18
25 18 0
9 4 5
19 6 9
9 2 5
% % %
9 2 0
3 2 0
6 2 0
6 6 18
3 10 0
3 0 14
12 4 0
Symptom
A vs. B A vs. C1 ⫹ C2 A vs. B ⫹ C1 ⫹ C2 B vs. C1 ⫹ C2
Discomfort Photophobia Discomfort Photophobia Discomfort Photophobia Lacrimation No significant results
Fundus Examinations No clinical evidence of cystoid macular edema or other posterior segment abnormality was noted during the study period.
Discussion This study represents our second human clinical trial using the Surodex Drug Delivery System to deliver a controlled intraocular dosage of dexamethasone after cataract extraction and confirms that, as in the first trial, placement of Surodex results in significant reduction in aqueous flare. This study was also able to establish that symptoms related to postoperative inflammation also were reduced significantly with Surodex, as compared with conventional 0.1% dexamethasone eyedrops. Additional evidence supporting the enhanced inflammatory control in Surodex-treated eyes is shown by the fact that 9 eyes (27%) in the eyedrops group required steroid augmentation (clinical therapeutic failures), whereas only 2 Surodex-treated eyes required additional steroids. However, the antiinflammatory efficacy of two Surodex pellets does not appear to enhance aqueous flare reduction, because there was no significant difference in flare values between one Surodex AC placement in the first trial and two Surodex AC placement in this study.
Pearson’s Statistic
Significance by Fisher Exact Test
14.94 7.49 16.67 7.56 22.77 5.18 7.09
0.00 0.01 0.00 0.01 0.00 0.04 0.01
It should be noted that, as in the first study, no significant differences between all three treatment arms were found with regard to clinical slit-lamp assessment of cells and flare grading, but this is likely to be the result of poor sensitivity and specificity in grading postoperative inflammation at the slit lamp. Slit-lamp grading of aqueous flare and cells is a subjective and semiquantitative procedure, prone to intraobserver error (for example, aqueous flare was arbitrarily graded into absent, mild, moderate, or strong). In addition, cells in the AC after intraocular surgery are comprised of a variety of erythrocytes, iris pigment cells, inflammatory cells, and other intraocular cell debris, and therefore do not reflect purely inflammatory activity.4 The original grading systems by Hogan et al5 and others were devised primarily for the grading of inflammation in uveitis patients, where the presumption of floaters being largely inflammatory is more appropriate. In addition, all eyes in our study were Asian with deeply pigmented irides, which we observe clinically to shed large amounts of pigment during extracapsular cataract extraction. Comparison of data from our first Surodex trial (in which one Surodex was compared with steroid eyedrops) showed that no significant differences were found between eyedrop control groups (groups A), confirming the presence of good standardization of flare meter values between the two stud-
2179
Ophthalmology Volume 108, Number 12, December 2001 Table 5. Central Corneal Endothelial Cell Counts Central Corneal Endothelial Cell Counts
Preoperative Group A (n ⫽ 33) Group B (n ⫽ 49) Group A (n ⫽ 33) Group C1 ⫹ C2 (n ⫽ 22) Group B (n ⫽ 49) Group C1 ⫹ C2 (n ⫽ 22) 3 months after surgery Group A (n ⫽ 33) Group B (n ⫽ 48) Group A (n ⫽ 33) Group C1 ⫹ C2 (n ⫽ 22) Group B (n ⫽ 48) Group C1 ⫹ C2 (n ⫽ 22) 1 year after surgery Group A (n ⫽ 26) Group B (n ⫽ 37) Group A (n ⫽ 26) Group C1 ⫹ C2 (n ⫽ 13) Group B (n ⫽ 37) Group C1 ⫹ C2 (n ⫽ 13) Repeated measurements adjusted for baseline value
Mean (SD)
Median (Range)
2450 (312) 2605 (322) 2450 (312) 2585 (384) 2604 (322) 2585 (383)
2487 (1814–2994) 2624 (1923–3412) 2487 (1814–2994) 2538 (1930–3937) 2624 (1923–3412) 2538 (1930–3937)
2400 (263) 2566 (353) 2400 (263) 2494 (273) 2566 (353) 2494 (273)
2421 (1669–2865) 2561 (1795–3378) 2421 (1669–2865) 2481 (2020–3300) 2561 (1795–3378) 2481 (2020–3300)
2370 (418) 2550 (328) 2370 (418) 2480 (268) 2550 (328) 2480 (268)
2398 (1620–3174) 2475 (1901–3436) 2398 (1620–3174) 2506 (1915–3021) 2475 (1901–3436) 2506 (1915–3021)
Statistical Significance*
— — — P ⫽ 0.03 P ⫽ 0.30 P ⫽ 0.27 P ⫽ 0.13 P ⫽ 0.32 P ⫽ 0.72
Group
P-value Adjusted for Repeated Measurements
Post Hoc Tests†
Interaction‡
A vs. B vs. C1 vs. C2 A vs. B vs. C1 ⫹ C2 A vs. B ⫹ C
0.106 0.056 0.021
NS NS —
NS NS NS
NS ⫽ not significant. *Significance by Wilcoxon rank sum test. † ‡
Bonferroni. Greenhouse-Geiser.
ies. We also compared one Surodex in the AC (previous study’s group B) with two Surodex in the AC (current study group B), which also showed no statistical differences at all time points, suggesting that no increased antiinflammatory efficacy was achieved with one additional Surodex. This suggests that placement of an additional Surodex pellet after routine uncomplicated ECCE does not result in enhanced antiinflammatory efficacy. Laser flare photometry is a Food and Drug Administration-approved technique to measure aqueous flare objectively in the AC, and its reproducibility and objectivity as a method of aqueous flare measurement has been confirmed by El-Maghraby et al6 and Sawa et al.7 The laser flare meter is now an established, noninvasive research tool used to detect subtle changes in the blood–aqueous barrier after surgical procedures that include small incision cataract surgery, capsulotomy, laser trabeculoplasty, phakic IOL implantation, radial keratotomy, excimer laser corneal surgery, and penetrating keratoplasty.8 –13 It therefore is reasonable to assume that the differences in aqueous flare obtained with the flare meter are more accurate than our arbitrary clinical slit-lamp grading technique of flare. It should be stated, however, that increased aqueous flare from disruption of the blood–aqueous barrier is directly related to, but may not
2180
fully equate to, intraocular inflammation, which requires an inflammatory cellular component. This study has also ascertained that posterior placement of Surodex results in significant AC migration within 1 week of surgery, as evidenced by the fact that in only 42% of group C eyes did PC placement remain in situ, whereas in 7 eyes (19%), one of the two Surodex pellets had migrated into the inferior AC angle, and in 14 eyes (39%), both Surodex pellets had migrated into the AC. In contrast, no migration was observed with AC placement. It should be noted that in eyes with PC placement, Surodex was placed preferentially within the ciliary sulcus, as opposed to in-thebag placement, to reduce the risk of pellet loculation within the capsular bag, and this may have resulted in a higher instability rate. In addition, deep placement into the depths of the ciliary sulcus was not achieved in most cases because of poor visibility and access behind the iris diaphragm. Although pellet migration occurs with PC placement, the overall efficacy of Surodex does not appear to depend on placement site, because this study was unable to show any significant differences in aqueous flare reduction between AC and PC placement. This suggests that exact localization of Surodex within the anterior segment may not be necessary to achieve adequate antiinflammatory efficacy. This
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery has importance in instances of poor anterior segment visualization. In addition, pellet migration within the visual axis does not seem to cause visual symptoms, as evidenced by the two group C patients in whom Surodex had migrated into the visual axis, between the IOL and the posterior capsule. Neither patient reported visual degradation on direct questioning. This study also reinforces previous safety findings of the first trial regarding ocular insertion of Surodex. Placement of two Surodex pellets either in the AC or in the ciliary sulcus of the PC did not result in significant intraocular complications. In particular, only one eye within the Surodex arms had an initial transient IOP rise, whereas minimal irritative or local effects related to Surodex insertion were seen. Despite the fact that in almost one third of cases residual pellet remnants remained present at 1 year, no late IOP rise was encountered. In addition, migration of Surodex into the visual axis, which occurred in two group C eyes, did not result in any visual disturbance (Fig 5). All trial patients attained 6/12 vision at 3 months, with the exception of one patient with early posterior capsule opacification. Posterior capsule rupture with vitreous loss and necessitating AC IOL placement also does not appear to be a contraindication for Surodex insertion. Corneal endothelial cell counts suggest that two Surodex inserted in the AC do not result in significant endothelial cell damage. In addition, the finding that no significant differences in mean inferior endothelial cells counts were encountered between group A and group B patients suggests that insertion of two Surodex pellets in the inferior angle of the AC does not lead to adjacent inferior endothelial cell damage or loss. These results are similar to the previous clinical trial with one Surodex insertion. In summary, this study demonstrates that placement of two Oculex Drug Delivery System devices containing dexamethasone in the eye after intraocular surgery is safe. Two Surodex pellets are equally effective, but not more effective than one Surodex, in reducing aqueous flare after cataract surgery as measured by laser flare photometry, and although anterior migration occurs with PC placement of the DDS, no difference in efficacy occurs. Surodex is currently approved for clinical use in Singapore, China, and Mexico, and phase III Food and Drug Administration clinical trials have been completed in the United States. Clinical trials on an antibiotic DDS are currently underway, and it is anticipated that the reality of cataract surgery without the need for eyedrops is close at hand. The use of Oculex DDS technology for controlled drug delivery in many other forms of ocular surgery and ocular diseases is also being explored.
Acknowledgment: The authors thank Professor Ng Tze Pin, Department of Community, Occupational and Family Medicine, National University of Singapore, for statistical consultation.
References 1. Munton CGF, Phillips CI, Martin B, et al. Vicryl (Polyglactin 910). A new synthetic absorbable suture in ophthalmic surgery. Br J Ophthalmol 1974;58:941–7. 2. Tan DTH, Chee SP, Lim L, Lim ASM. Randomized clinical trial of a new dexamethasone delivery system (Surodex) for treatment of post-cataract surgery inflammation. Ophthalmology 1999;106:2:223–31. 3. Kraff MC, Martin RG, Neumann AC, Weinstein AJ. Efficacy of diclofenac sodium ophthalmic solution versus placebo in reducing inflammation following cataract extraction and posterior chamber lens implantation. J Cataract Refract Surg 1994;20:138 – 44. 4. Mishima S, Tanishima T, Masuda K. Pathophysiology and pharmacology of intraocular surgery. Aust N Z J Ophthalmol 1985;13:147–58. 5. Hogan MJ, Kimura SJ, Thygeson P. Signs and symptoms of uveitis. I. Anterior uveitis. Am J Ophthalmol 1959;47(5 pt 1):155–76. 6. El-Maghraby A, Marzouki A, Matheen TM, et al. Reproducibility and validity of laser flare/cell meter measurements of intraocular inflammation. J Cataract Refract Surg 1993;19: 52–5. 7. Sawa M, Tsurimaki Y, Tsuru T, Shimizu H. New quantitative method to determine protein concentration and cell number in aqueous in vivo. Jpn J Ophthalmol 1988;32:132– 42. 8. Martin RG, Sanders DR, Van Der Karr MA, DeLuca M. Effect of small incision intraocular lens surgery on postoperative inflammation and astigmatism. A study of the AMO SI-18NB small incision lens. J Cataract Refract Surg 1992; 18:51–7. 9. Altamirano D, Mermoud A, Pittet N, et al. Aqueous humor analysis after Nd:YAG laser capsulotomy with the laser flarecell meter. J Cataract Refract Surg 1992;18:554 – 8. 10. Alio´ JL, de la Hoz F, Ismail MM. Subclinical inflammatory reaction induced by phakic anterior chamber lenses for the correction of high myopia. Ocular Immunol Inflamm 1993;1: 219 –23. 11. Vita RCA, Campos M, Belfort R Jr, Paiva ER. Alterations in blood–aqueous barrier after corneal refractive surgery. Cornea 1998;17:158 – 62. 12. Ku¨ chle M, Nguyen NX, Naumann GOH. Aqueous flare following penetrating keratoplasty and in corneal graft rejection. Arch Ophthalmol 1994;112:354 – 8. 13. Ku¨ chle M, Nguyen NX, Seitz B, et al. Blood–aqueous barrier after mechanical or nonmechanical excimer laser trephination in penetrating keratoplasty. Am J Ophthalmol 1998;125:177– 81.
2181
2172
© 2001 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
material in general and ocular surgery as Vicryl (Johnson & Johnson Medical Inc., Jurong, Singapore) has been used, and the polymer has been shown to degrade safely by hydrolysis into the natural byproducts of lactic acid and glycolic acid, which are metabolized.1 The potential advantages of DDS over conventional postoperative eyedrops are
Supported by the Singapore National Medical Research Council (grant no.: STB/0050/1991), Singapore, and Oculex Pharmaceuticals, Inc., Sunnyvale, California. The authors do not have a financial interest in this or competing products. Correspondence to Donald T. H. Tan, FRCS, FRCOphth, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 0316. E-mail: snecdt@ pacific.net.sg ISSN 0161-6420/01/$–see front matter PII S0161-6420(01)00839-9
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery (1) significantly lower drug concentrations required (and hence reduced systemic side effects and systemic drug toxicity), (2) controlled anterior chamber (AC) drug concentrations with near-zero order pharmacokinetics, and (3) reduced complications related to patient noncompliance with eyedrop administration. Surodex is a DDS formulation containing 60 g of dexamethasone designed to reduce postoperative inflammation after cataract surgery. Surodex has previously been shown to be both safe and effective in delivering dexamethasone into animal as well as human eyes. Kochinke et al [Invest Ophthalmol Vis Sci 35(Suppl):4474, 1994] showed that Surodex provided effective controlled drug delivery in the AC with near-zero order pharmacokinetics over a 7-day period in rabbit eyes. Wong et al also demonstrated the safety of Surodex in the AC in rabbits in a Food and Drug Administration phase I toxicology study [FDA study protocol no. 94001], and Surodex was initially implanted in six human eyes undergoing phacoemulsification in a Food and Drug Administration phase I pilot human eye trial [FDA study protocol no. C95-001]. We previously performed a randomized, partially controlled human clinical trial comparing Surodex with conventional 0.1% dexamethasone eyedrops in cataract patients undergoing extracapsular cataract extraction (ECCE) with intraocular lens (IOL) implantation.2 Results showed that placement of one Surodex in the inferior angle of the AC was significantly more effective in reducing aqueous flare levels after surgery, as measured with laser flare photometry for up to 30 days after surgery. In addition, Surodex was found to be safe, with no complications occurring within the trial, no local reaction around the site of Surodex placement, and no significant difference in corneal endothelial cell counts between Surodex and eyedrop treatment arms 1 year after surgery. The aim of this clinical trial was to evaluate the safety and antiinflammatory efficacy of two DDS pellets of Surodex in the eye, as compared with conventional steroid eyedrops, and to evaluate the relative safety and efficacy of Surodex placement in the AC versus the posterior chamber (PC).
Materials and Methods One hundred four patients undergoing elective ECCE with IOL implantation were enrolled in a prospective, double-masked, randomized, partially controlled clinical trial during the period of August 1996 through January 1998 at the Singapore National Eye Centre, the tertiary referral national ophthalmic institute in Singapore. The clinical trial protocol and informed consent form were approved by the Singapore National Eye Centre Ethics and Research Committees, and written informed consent was obtained. Study conduct conformed to scientific principles embodied in the World Medical Association Declaration of Helsinki, as revised in 1989. Inclusion criteria included consecutive patients with uncomplicated, senile cataract listed for ECCE and IOL surgery by the two study consultant surgeons (D.T.H.T., S.P.C.). Exclusion criteria included patients taking oral or topical antiinflammatory agents, with diabetes mellitus, with a history of steroid-induced ocular hypertension, with hypermature cataracts, and eyes with uveitis, glaucoma, and corneal disease. Only one eye of each patient was enrolled in the trial.
Treatment Assignation and Study Masking Study participants were stratified into three categories according to age (50 –59 years, 60 – 69 years, and 70 –79 years). Eyes within each category were assigned randomly to receive one of three postoperative steroid treatment regimens from a table of computergenerated random numbers. Group A patients received conventional 0.1% dexamethasone eyedrops four times daily for a period of 30 days after surgery. Group B patients received two Surodex pellets inserted into the inferior angle of the AC at the conclusion of surgery and were prescribed normal saline placebo eyedrops. Group C patients received two Surodex pellets inserted into the inferior ciliary sulcus of the PC at the conclusion of surgery and were prescribed normal saline placebo eyedrops. Patients and study investigators were masked as to treatment regimens and, to reduce surgical bias, treatment groups were assigned by a study nurse coordinator who was the only person during the trial who was aware of patients’ treatment group status. The study code was kept in a locked drawer by the study nurse coordinator in the center’s clinical trials office. Surgeons were only informed of individual patients’ group status immediately before wound closure, at the conclusion of surgery, when Surodex insertion was required. After surgery, to reduce observer bias, all clinical assessment of intraocular inflammation was performed by a separate, masked investigator or ophthalmic technician who was unaware of patients’ group status. No inadvertent unmasking of patients occurred during the study.
Statistical Analysis and Sample Size Calculation The primary endpoint to compare relative efficacy of two Surodex pellets either in the AC (group B) or PC (group C) versus steroid drops was defined as the degree of AC flare, as measured objectively by the laser flare meter at the preset time points of 4, 8, 15, 30, and 90 days after surgery. A secondary outcome measure was chosen to be the semiquantitative measurement of cells and flare by slit-lamp grading. Treatment comparisons were performed with two-tailed significance levels set at P ⬍ 0.025 (Bonferroni-Holm) for all pair-wise comparisons, i.e., combined groups B and C versus group A, group B versus group A, and group C versus group A. The degree of inflammation is a continuous variable, and the Wilcoxon signed-rank tests of significance for expected skewed distribution of data were used for analysis of results. Both parametric and nonparametric tests were performed and showed consistent results. A secondary analysis of the data was performed with repeated measures analysis of variance with borderline covariate adjustment. This was performed for analysis of laser flare meter values and for endothelial cell counts. Study results were also compared directly with those obtained in our previous study with one Surodex insertion, which had a similar study methodology. All results were analyzed initially on an intention-to-treat basis. However, as the study progressed, it became clear that significant pellet migration to the AC was noted when Surodex was placed in the ciliary sulcus of the PC (group C), and gonioscopy at day 8 revealed that of the 36 eyes in group C, 15 eyes (58%) had evidence of either one or two Surodex present in the AC. With this in mind, a reclassification of group C was performed based on gonioscopic status at day 8: group C1 denoted eyes with evidence of one Surodex in the AC (and therefore one additional Surodex in the PC), whereas those group C eyes with no Surodex seen on day 8 gonioscopy (i.e., no pellet migration occurred) were categorized as group C2. Those eyes with two Surodex migrated to the AC were recategorized as group B. All study analyses conformed to this recategorization of treatment groups. Sample size calculations were based on available data on laser flare meter and slit-lamp inflammatory grading results accrued
2173
Ophthalmology Volume 108, Number 12, December 2001 from our previous trial using one Surodex. For laser flare meter values, to demonstrate statistical significance with a type I error (␣) of 0.025 and power of 80%, for an expected difference of 15 photons ms⫺1 (standard deviation, 15.0), the minimal sample size per treatment group (1:1:1 allocation) was 20 patients. For slitlamp grading, to demonstrate statistical significance with a type I error (␣) of 0.05 and power of 80%, for an expected difference of 1.0 (standard deviation, 1.5), the minimal sample size was 36 patients.
Definition of Clinical Therapeutic Failures Cases requiring additional use of antiinflammatory agents outside the study regimen at any postoperative period were considered as clinical therapeutic failures. Judgment was based on a subjective clinical evaluation of the degree of ocular inflammation present at either the ocular surface, limbal wound, or intraocularly. This clinical decision to augment steroid therapy was made by the study surgeons and was independent of the study’s postoperative investigator inflammatory grading.
Preoperative Assessment Before surgery, unaided and aided Snellen visual acuity, manifest refraction, Goldmann tonometry, and slit-lamp and fundus examinations were performed in all cases. Preoperative endothelial specular microscopy with endothelial cell morphometric analysis of the central and inferior corneal endothelium was performed with the Konan NONCON ROBO noncontact specular microscope (Konan, Osaka, Japan). Preoperative laser flare meter measurements were performed with the Kowa FC500 Laser Flare Meter (Kowa Co. Ltd., Tokyo, Japan).
Laser Flare Meter Measurement Technique All eyes were dilated with one drop of tropicamide 1% eyedrop before laser flare meter measurement, and all measurements were performed within 30 minutes of instillation. Laser flare measurement was performed in a totally dark room by a trained technician who was masked to patient group status, and a minimum of 10 recorded flare values were averaged for each visit.
Preoperative Medication All eyes were dilated before surgery with two drops of phenylephrine 10% eyedrop administered 5 minutes apart, three drops of tropicamide 1% eyedrop administered 3 minutes apart, and six drops of flurbiprofen sodium 0.03% eyedrop, administered 10 minutes apart. Preoperative antibiotic prophylaxis consisted of one drop of chloramphenicol 0.5% eyedrop administered 1 hour before surgery.
Surgical Procedure and Insertion of Surodex All cases received either retrobulbar or peribulbar anesthesia. Extracapsular cataract extraction was performed through a superior limbal incision, using a conventional can-opener capsulotomy technique, and the nucleus was delivered under Healon. Manual irrigation-aspiration was carried out with a Simcoe cannula, and insertion of a one-piece polymethyl methacrylate IOL (Cilco MZ60BD) was performed in all cases. After IOL insertion, the study envelope was opened by the operating theater circulating nurse to ascertain group status of the patient. For group B patients, two Surodex pellets were inserted into the AC with the use of Kelman-McPherson forceps and gently positioned in the inferior
2174
angle with the help of a Sinskey hook (Figs 1 and 2). For group C patients, two Surodex pellets were inserted into the ciliary sulcus of the PC with the use of Kelman-McPherson forceps. A Sinskey hook was used to push the pellets inferiorly out of sight beneath the iris. Wound closure was then effected with the use of 6 to 7 interrupted 10-0 nylon sutures, and Healon was carefully aspirated from the AC. A subconjunctival injection of gentamicin (4 mg) completed the procedure. All surgery was videotaped and archived.
Postoperative Medication All patients were prescribed chloramphenicol 0.5% eyedrop four times daily for 30 days, and oral acetazolamide 500 mg immediately after surgery and 250 mg twice daily for 48 hours after surgery. Patients were prescribed masked study eyedrops by the study nursing coordinator on the day of surgery and were instructed to apply four drops per day at equal intervals for 30 days. Study eyedrops for group A patients consisted of dexamethasone 0.1% eyedrop (Gutt Maxidex), whereas group B and C patients were prescribed 0.9% sodium chloride eyedrops (Xepa-Soul Pattinson (M) Sendirian Berhad) as placebo. Identical eyedrop bottles and labeling were used for all groups, and the surgeon, postoperative investigator, and patients were masked as to the study eyedrops.
Postoperative Evaluation Patients initially were requested to return for follow-up visits on days 1, 2, 4, 8, 15, 30, 60, and 90 after surgery. Subsequent to journal reviewers’ request for longer follow-up for our first Surodex study, it was decided that patients should also be followed up for at least 12 months, and the additional 12-month visit was added. Attempts also were made to recall all patients who had already been seen for their final 90-day visit. Examination at each postoperative visit included unaided and aided Snellen visual acuity, Goldmann applanation tonometry, and slit-lamp examination. Laser flare meter measurement was performed at each visit up to the day 90 visit. A separate, masked investigator performed slit-lamp grading of flare and cells. Gonioscopy was initially performed on the fourth postoperative day and at every subsequent postoperative visit (Figs 3, 4, 5, and 6). At the 90-day visit, fundus examination and manifest refraction were performed in all cases. Central and inferior corneal endothelial specular microscopy with endothelial cell morphometric analysis was performed after surgery at the 90-day visit and at the 1-year visit.
Clinical Assessment of Ocular Inflammation At each visit, ocular inflammation was assessed clinically by direct questioning for symptoms of inflammation (the presence of ocular discomfort or pain, tearing, and photophobia), as well as slit-lamp grading of intraocular cells and flare, conjunctival hyperemia, and ciliary injection according to established grading scales previously described by Kraff et al.3 Anterior chamber cell and flare grading scores were added together as a combined inflammatory score. Particular attention was paid to masking of the investigator grading inflammation, which was achievable because the technique of cell and flare assessment that involved the use of a minimum slit aperture of 0.2 mm and centration of the slit beam in the AC precluded a view of the inferior AC.
Results One hundred four eyes of 104 patients were enrolled in the study from August 1996 through January 1998, with 33 eyes in group A,
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery
Figure 1. Slit-lamp photograph of an isolated instance in which two Surodex are visible in the inferior angle. Note that in most such cases, Surodex pellets were sited deeper within the angle and were not visible on slit-lamp examination. Figure 2. Gonioscopic view of inferior angle shows two Surodex placed adjacent to each other in the inferior angle. Figure 3. Gonioscopic view of inferior angle showing Surodex remnants in early dissolution. Note that the peripheral cornea adjacent to Surodex remnants remains clear, and the inferior angle appears normal with no evidence of peripheral synechiae or scarring. Figure 4. Gonioscopic view of inferior angle showing Surodex remnants in late dissolution (6 months). Minute remnants or residue remains in the angle. Figure 5. Gonioscopic view showing normal inferior angle structures after complete Surodex dissolution. Figure 6. Slit-lamp photograph showing mild globular, lobulated appearance of dissolving Surodex. No adjacent corneal damage or clouding is seen.
2175
Ophthalmology Volume 108, Number 12, December 2001 Table 1. Mean Laser Flare Meter Baseline Values: Repeated Measurements Adjusted for Baseline Value
Group
P Value for Repeated Measurements
Post Hoc Tests*
Interaction† Time* Group
A vs. B vs. C1 vs. C2 A vs. B vs. C1 ⫹ C2 A vs. B ⫹ C
0.053 0.025 0.007
NS A vs. B, P ⫽ 0.027 —
NS 0.033 0.008
NS ⫽ not significant. *Bonferroni. † Greenhouse-Geiser.
35 eyes in group B, and 36 eyes in group C. The mean ages of patients in groups A, B, and C were 67.4 years, 66.4 years, and 68.3 years, respectively. Forty-six percent of patients were male and 54% were female. Most patients were Chinese (85%), whereas 7%, 4%, and 4% of patients were Malays, Indians, and other racial groups, respectively. This approximated the racial mix in Singapore. There were no statistically significant differences in age, race, or gender in the study groups. All 104 patients completed the initial study exit period at 90 days, but only 87 patients (83.7%) attended the 1-year recall visit. Eighty-five percent of group A patients, 82% of group B patients, 100% of group C1 patients and 80% of group C2 patients completed the additional 1-year visit. Comparison of visual acuity, IOP, symptoms, and endothelial cell counts obtained during the last study visit between the 17 patients who did not complete the month 12 follow-up and the other 87 patients revealed no statistically significant differences between the two groups.
Laser Flare Meter Results Mean laser flare meter values for individual study arms were analyzed at all time points up to 3 months after surgery. Preoperative flare values showed a difference between treatment groups, with generally higher preoperative flare values in the Surodex treated eyes. This was confirmed with repeated measurements, including baseline flare value as a covariant (Table 1). After surgery, mean flare values were lower in the Surodex groups (B and C) as compared with the eyedrops group (A) at all visits up to day 90 (Table 2; Fig 7). Flare reduction in the Surodex groups attained statistical significance at days 4, 8, and 15 (P ⫽ 0.001, P ⫽ 0.001, P ⫽ 0.02, respectively). When groups B and C were compared individually with group A, statistical significance was also achieved at these same time points. However, when groups C1 and C2 were compared with group A, no statistical significance was achieved in the group C1, whereas statistical significance was achieved in group C2 at days 4, 15, 30, and 60. Analysis of mean flare values between Surodex groups (B, C1, and C2) showed no statistically significant difference in efficacy between AC or PC placement of Surodex (Table 2).
Clinical Inflammation and Symptom Scores No significant differences in slit-lamp combined cell and flare scores occurred between the three treatment arms at any study time point (Table 3; Fig 8). No differences in ciliary injection or hyperemia were noted between treatment arms (data not shown). However, evaluation of inflammatory symptoms showed a clear difference between the eyedrop-treated group and Surodex groups for discomfort, photophobia, and lacrimation (Table 4). The percentage of patients who reported these symptoms at individual
2176
visits was significantly higher in the eyedrop-treated group compared with the Surodex-treated groups.
Dissolution of Surodex As in our previous study, we noted that the Surodex pellets gradually dissolved into ovoid or circular remnants at the site of original insertion and finally disappeared completely or left tiny residues (Figs 3, 4, and 5). In some instances, a lobulated, semitranslucent change occurred, associated with a slight increase in size, possibly suggesting pellet hydration (Fig 6). At the day 90 visit, gonioscopy confirmed that 57 eyes of the total of 71 eyes with at least one Surodex originally observed in the AC angle (80%) still had evidence of pellet remnants in the angle. At 6 months, pellet remnants were noted to be less, with 41 eyes (57%) still exhibiting remnants in the inferior angle, whereas at 1 year, 22 eyes (31%) had minute residual pellet remnants or residues.
Localized Angle Changes As in our previous study, mild localized angle changes were noted at or adjacent to pellet remnants in a percentage of eyes. In all cases, these were mild degrees of focal peripheral anterior synechiae with a notched appearance that were not progressive and did not result in pupil distortion. These lesions were visible only on careful gonioscopic examination, were attributed to focal adherence of iris to the DDS pellet, and were deemed clinically insignificant. Of the 71 eyes with Surodex present in the angle, 22 eyes (31%) exhibited these mild focal peripheral anterior synechiae lesions.
Corneal Endothelial Cell Morphometric Analysis No reduction in endothelial cell count occurred in Surodex-treated eyes, as compared with the dexamethasone eyedrop-treated eyes (Table 5). Because the baseline value was associated significantly with follow-up values, repeated-measures analysis of variance with borderline covariate adjustment was performed, with post hoc testing confirming a lack of any significant difference (Table 1). Similar analysis for inferior endothelial cell count also showed no significant differences between treatment arms (data not shown). In addition, the percentage change in 3-month and 1-year postoperative central and inferior endothelial cell count values from preoperative baseline values was also compared (data not shown). No significant differences in percentage change of cell counts were noted among all three treatment arms at both the 90-day visit and the 1-year visit.
Therapeutic Failures There were 9 eyes in group A (27%) that clinically required steroid augmentation because of excessive postoperative inflammation or
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery Table 2. Mean Laser Flare Meter Values: Comparison between Eyedrops and Surodex Groups and Position of Surodex Preoperative
Day 4
Day 8
Day 15
Day 30
Day 60
Day 90
Group A N ⫽ 33 N ⫽ 32 N ⫽ 32 Mean 6.15 33.99 24.52 SD 2.3 34.1 16.3 Median 5.6 24.3 18.0 Group B N ⫽ 49 N ⫽ 46 N ⫽ 46 Mean 6.27 20.19 16.68 SD 1.4 7.5 8.5 Median 6.2 18.2 14.3 Group B ⫹ C N ⫽ 71 N ⫽ 67 N ⫽ 65 Mean 6.35 20.75 17.79 SD 1.7 10.2 12.7 Median 6.1 18.2 14.2 Group C1 N⫽7 N⫽6 N⫽6 Mean 7.64 18.18 14.60 SD 3.2 3.7 4.9 Median 6.0 18.3 12.8 Group C2 N ⫽ 15 N ⫽ 15 N ⫽ 13 Mean 6.00 23.47 23.20 SD 1.4 17.0 23.2 Median 5.8 18.2 14.4 Group C1 ⫹ C2 N ⫽ 22 N ⫽ 21 N ⫽ 19 Mean 6.52 21.96 20.48 SD 2.2 14.6 19.6 Median 5.9 18.2 14.2 Significance levels for comparison between eyedrops and Surodex groups A vs. B ⫹ C 0.00 0.00 A vs. B 0.01 0.00 A vs. C1 ⫹ C2 0.01 0.04 A vs. C1 0.21 0.26 A vs. C2 0.04 0.21 Significance levels for comparison between anterior and posterior positions B vs. C2 0.86 0.65 B vs. C1 0.81 0.59 B vs. C1 ⫹ C2 0.80 0.94
N ⫽ 32 21.38 22.4 16.7 N ⫽ 49 13.47 5.5 12.7 N ⫽ 71 13.50 6.1 12.5 N⫽7 13.81 5.9 13.3 N ⫽ 15 13.33 8.3 11.1 N ⫽ 22 13.49 7.4 11.7
N ⫽ 32 12.78 6.1 11.6 N ⫽ 48 10.32 3.9 9.2 N ⫽ 70 10.14 4.0 9.0 N⫽7 10.31 3.1 10.0 N ⫽ 15 9.47 5.0 8.0 N ⫽ 22 9.74 4.4 8.1
N ⫽ 31 11.19 4.6 10.1 N ⫽ 48 9.69 3.7 9.2 N ⫽ 70 9.36 3.5 8.9 N⫽7 9.27 1.6 9.7 N ⫽ 15 8.25 3.1 7.3 N ⫽ 22 8.57 2.7 8.1
N ⫽ 33 9.09 3.5 8.8 N ⫽ 47 8.86 2.4 8.8 N ⫽ 69 8.70 2.3 8.7 N⫽7 8.71 1.7 8.5 N ⫽ 15 8.19 1.9 8.3 N ⫽ 22 8.36 1.9 8.4
0.02 0.04 0.04 0.34 0.04
0.07 0.15 0.05 0.46 0.04
0.05 0.16 0.03 0.53 0.02
0.88 0.76 0.86 0.85 0.72
0.43 0.96 0.56
0.16 0.85 0.31
0.09 0.67 0.26
0.31 0.96 0.41
SD ⫽ standard deviation. P values in bold type are statistically significant (P ⬍ 0.05).
significant remaining intraocular inflammation at 1 month, and therefore were clinical therapeutic failures. In contrast, only two eyes among the Surodex-treated eyes (one in group B, one in group C) were clinical therapeutic failures. This reduction in therapeutic failures in Surodex-treated eyes as compared with eyedrop-treated
eyes was statistically significant (P ⫽ 0.001, Pearson’s chi-square test). No sight-threatening complications occurred in these eyes with therapeutic failures.
Intraoperative Complications Inferior zonulysis with vitreous loss occurred in one group B patient. Anterior vitrectomy with AC IOL implanted successfully, and Surodex placements in the inferior chamber angle were performed. After surgery, no complications were encountered in this patient, and visual acuity attained was 6/6. No other intraoperative complications were encountered.
Postoperative Complications
Figure 7. Line graph showing mean laser flare meter values for all groups.
No severe complications were encountered after surgery. In group A, mild complications included transient raised IOP in two eyes (one requiring a short course of topical -blockers), blepharitis (n ⫽ 1), keratitis medicamentosa (n ⫽ 1), and fibrotic posterior capsule opacification in two eyes, with visual acuities of 6/12 and 6/18 (neodymium:yytrium–aluminum– garnet laser capsulotomy not required). One group A patient was noted to have excessive inflammation at the wound margin, together with a localized scleritis, which responded well to a 1-week course of oral prednisolone. In group B, mild complications included a mild surgical hyphema clot in one patient on day 1, which resolved, corneal
2177
Ophthalmology Volume 108, Number 12, December 2001 Table 3. Combined Cell and Flare Slit-lamp Grading Preoperative
Day 4
Day 8
Day 15
Cell grading (slit-lamp) Group A N ⫽ 33 N ⫽ 33 N ⫽ 33 N ⫽ 32 Mean 0 2.76 2.21 1.66 SD 0 1.0 0.8 0.7 Median 0 3 2 2 Group B ⫹ C N ⫽ 71 N ⫽ 71 N ⫽ 71 N ⫽ 71 Mean 0.01 3.13 2.24 1.63 SD 0.1 0.9 1.0 0.6 Median 0 3 2 2 Flare grading (slit-lamp) Group A N ⫽ 33 N ⫽ 33 N ⫽ 33 N ⫽ 32 Mean 0 1.27 0.97 0.53 SD 0 0.5 0.6 0.7 Median 0 1 1 0 Group B ⫹ C N ⫽ 71 N ⫽ 71 N ⫽ 71 N ⫽ 71 Mean 0.03 1.13 0.89 0.48 SD 0.2 0.6 0.6 0.6 Median 0 1 1 0 Combined cell and flare Group A N ⫽ 33 N ⫽ 33 N ⫽ 33 N ⫽ 32 Mean 0 4.03 3.18 2.19 SD 0 1.3 1.2 1.0 Median 0 4 3 2 Group B ⫹ C N ⫽ 71 N ⫽ 71 N ⫽ 71 N ⫽ 71 Mean 0.07 4.21 3.13 2.11 SD 0.3 1.3 1.4 1.0 Median 0 4 3 2 Significance levels for comparison between eyedrops and Surodex groups (A vs. B ⫹ C) Cell 0.06 0.98 0.85 Flare 0.22 0.51 0.74 Combined 0.26 0.97 0.70
Day 30
Day 60
Day 90
N ⫽ 32 0.91 0.8 1 N ⫽ 70 0.97 0.7 1
N ⫽ 31 0.84 0.7 1 N ⫽ 70 0.70 0.6 1
N ⫽ 32 0.50 1.0 0 N ⫽ 69 0.51 0.6 0
N ⫽ 32 0.19 0.4 0 N ⫽ 70 0.19 0.4 0
N ⫽ 31 0.13 0.3 0 N ⫽ 70 0.11 0.3 0
N ⫽ 32 0.06 0.3 0 N ⫽ 69 0.04 0.2 0
N ⫽ 32 1.09 1.0 1 N ⫽ 70 1.14 0.9 1
N ⫽ 31 0.97 0.8 1 N ⫽ 70 0.77 0.7 1
N ⫽ 32 0.36 0.6 0 N ⫽ 69 0.55 0.7 0
0.61 0.98 0.64
0.37 0.83 0.22
0.49 0.70 0.15
SD ⫽ standard deviation. Note: Wilcoxon rank sum test.
abrasion (n ⫽ 1), lower eyelid entropion (n ⫽ 1), and blepharitis (n ⫽ 1). In group C, mild complications included a mild surgical hyphema clot in one patient on day 1, which resolved, a peaked pupil resulting from an anterior capsule tag, raised IOP in one eye that resolved with a short course of topical -blockers, and blepharitis in two eyes. In addition, in two eyes, pellet migration into the
visual axis occurred between the posterior convexity of the IOL and the posterior capsule. However, pellet remnants dissolved without residual posterior capsule opacification, and no visual complaints or disturbances were noted by either patient on direct questioning. Peripheral anterior synechiae at the site of pellet implantation in the inferior chamber angle was noted in 18 of 71 eyes (25%) with Surodex in the AC angle. In all cases, the area of synechiae was minimal, involving less than one-half a clock hour of angle, and in all cases this was deemed not clinically significant. No progression of the degree of synechiae was noted, and no complications arose from these areas of synechiae.
Visual Acuity
Figure 8. Line graph showing mean combined cell and flare clinical slit-lamp results between treatment arms.
2178
At day 90, all eyes attained 6/12 or better corrected or uncorrected visual acuity, with the exception of one eye from group A that attained a best-corrected visual acuity of 6/18 because of the presence of mild fibrotic posterior capsule opacification. No loss of visual acuity attributable to Surodex insertion was seen in groups B or C. At the 1-year follow-up, two eyes (one in group B and one in group C) were noted to have foveal changes consistent with age-related macular degeneration, and attained best-corrected visual acuity of 6/18 and 6/15, respectively. In addition, one group A patient attained best-corrected visual acuity of 6/18, attributed to a combination of asteroid hyalosis and high corneal astigmatism of 5 diopters. No loss of acuity was related to Surodex insertion.
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery Table 4. Inflammation Symptoms: Percentage within Groups Day
Discomfort A B C1 ⫹ C2 Pain A B C1 ⫹ C2 Photophobia A B C1 ⫹ C2 Lacrimation A B C1 ⫹ C2 Others A B C1 ⫹ C2 Significance levels
1
4
8
15
30
60
90
% % %
24 4 5
27 20 9
21 10 5
25 12 9
19 4 9
16 6 0
3 0 0
% % %
3 0 5
3 0 0
0 0 0
6 2 0
3 0 0
0 2 0
0 0 0
% % %
6 0 5
3 0 0
3 0 5
3 0 0
0 0 0
0 0 0
0 0 0
% % %
21 8 9
24 27 23
33 31 18
25 18 0
9 4 5
19 6 9
9 2 5
% % %
9 2 0
3 2 0
6 2 0
6 6 18
3 10 0
3 0 14
12 4 0
Symptom
A vs. B A vs. C1 ⫹ C2 A vs. B ⫹ C1 ⫹ C2 B vs. C1 ⫹ C2
Discomfort Photophobia Discomfort Photophobia Discomfort Photophobia Lacrimation No significant results
Fundus Examinations No clinical evidence of cystoid macular edema or other posterior segment abnormality was noted during the study period.
Discussion This study represents our second human clinical trial using the Surodex Drug Delivery System to deliver a controlled intraocular dosage of dexamethasone after cataract extraction and confirms that, as in the first trial, placement of Surodex results in significant reduction in aqueous flare. This study was also able to establish that symptoms related to postoperative inflammation also were reduced significantly with Surodex, as compared with conventional 0.1% dexamethasone eyedrops. Additional evidence supporting the enhanced inflammatory control in Surodex-treated eyes is shown by the fact that 9 eyes (27%) in the eyedrops group required steroid augmentation (clinical therapeutic failures), whereas only 2 Surodex-treated eyes required additional steroids. However, the antiinflammatory efficacy of two Surodex pellets does not appear to enhance aqueous flare reduction, because there was no significant difference in flare values between one Surodex AC placement in the first trial and two Surodex AC placement in this study.
Pearson’s Statistic
Significance by Fisher Exact Test
14.94 7.49 16.67 7.56 22.77 5.18 7.09
0.00 0.01 0.00 0.01 0.00 0.04 0.01
It should be noted that, as in the first study, no significant differences between all three treatment arms were found with regard to clinical slit-lamp assessment of cells and flare grading, but this is likely to be the result of poor sensitivity and specificity in grading postoperative inflammation at the slit lamp. Slit-lamp grading of aqueous flare and cells is a subjective and semiquantitative procedure, prone to intraobserver error (for example, aqueous flare was arbitrarily graded into absent, mild, moderate, or strong). In addition, cells in the AC after intraocular surgery are comprised of a variety of erythrocytes, iris pigment cells, inflammatory cells, and other intraocular cell debris, and therefore do not reflect purely inflammatory activity.4 The original grading systems by Hogan et al5 and others were devised primarily for the grading of inflammation in uveitis patients, where the presumption of floaters being largely inflammatory is more appropriate. In addition, all eyes in our study were Asian with deeply pigmented irides, which we observe clinically to shed large amounts of pigment during extracapsular cataract extraction. Comparison of data from our first Surodex trial (in which one Surodex was compared with steroid eyedrops) showed that no significant differences were found between eyedrop control groups (groups A), confirming the presence of good standardization of flare meter values between the two stud-
2179
Ophthalmology Volume 108, Number 12, December 2001 Table 5. Central Corneal Endothelial Cell Counts Central Corneal Endothelial Cell Counts
Preoperative Group A (n ⫽ 33) Group B (n ⫽ 49) Group A (n ⫽ 33) Group C1 ⫹ C2 (n ⫽ 22) Group B (n ⫽ 49) Group C1 ⫹ C2 (n ⫽ 22) 3 months after surgery Group A (n ⫽ 33) Group B (n ⫽ 48) Group A (n ⫽ 33) Group C1 ⫹ C2 (n ⫽ 22) Group B (n ⫽ 48) Group C1 ⫹ C2 (n ⫽ 22) 1 year after surgery Group A (n ⫽ 26) Group B (n ⫽ 37) Group A (n ⫽ 26) Group C1 ⫹ C2 (n ⫽ 13) Group B (n ⫽ 37) Group C1 ⫹ C2 (n ⫽ 13) Repeated measurements adjusted for baseline value
Mean (SD)
Median (Range)
2450 (312) 2605 (322) 2450 (312) 2585 (384) 2604 (322) 2585 (383)
2487 (1814–2994) 2624 (1923–3412) 2487 (1814–2994) 2538 (1930–3937) 2624 (1923–3412) 2538 (1930–3937)
2400 (263) 2566 (353) 2400 (263) 2494 (273) 2566 (353) 2494 (273)
2421 (1669–2865) 2561 (1795–3378) 2421 (1669–2865) 2481 (2020–3300) 2561 (1795–3378) 2481 (2020–3300)
2370 (418) 2550 (328) 2370 (418) 2480 (268) 2550 (328) 2480 (268)
2398 (1620–3174) 2475 (1901–3436) 2398 (1620–3174) 2506 (1915–3021) 2475 (1901–3436) 2506 (1915–3021)
Statistical Significance*
— — — P ⫽ 0.03 P ⫽ 0.30 P ⫽ 0.27 P ⫽ 0.13 P ⫽ 0.32 P ⫽ 0.72
Group
P-value Adjusted for Repeated Measurements
Post Hoc Tests†
Interaction‡
A vs. B vs. C1 vs. C2 A vs. B vs. C1 ⫹ C2 A vs. B ⫹ C
0.106 0.056 0.021
NS NS —
NS NS NS
NS ⫽ not significant. *Significance by Wilcoxon rank sum test. † ‡
Bonferroni. Greenhouse-Geiser.
ies. We also compared one Surodex in the AC (previous study’s group B) with two Surodex in the AC (current study group B), which also showed no statistical differences at all time points, suggesting that no increased antiinflammatory efficacy was achieved with one additional Surodex. This suggests that placement of an additional Surodex pellet after routine uncomplicated ECCE does not result in enhanced antiinflammatory efficacy. Laser flare photometry is a Food and Drug Administration-approved technique to measure aqueous flare objectively in the AC, and its reproducibility and objectivity as a method of aqueous flare measurement has been confirmed by El-Maghraby et al6 and Sawa et al.7 The laser flare meter is now an established, noninvasive research tool used to detect subtle changes in the blood–aqueous barrier after surgical procedures that include small incision cataract surgery, capsulotomy, laser trabeculoplasty, phakic IOL implantation, radial keratotomy, excimer laser corneal surgery, and penetrating keratoplasty.8 –13 It therefore is reasonable to assume that the differences in aqueous flare obtained with the flare meter are more accurate than our arbitrary clinical slit-lamp grading technique of flare. It should be stated, however, that increased aqueous flare from disruption of the blood–aqueous barrier is directly related to, but may not
2180
fully equate to, intraocular inflammation, which requires an inflammatory cellular component. This study has also ascertained that posterior placement of Surodex results in significant AC migration within 1 week of surgery, as evidenced by the fact that in only 42% of group C eyes did PC placement remain in situ, whereas in 7 eyes (19%), one of the two Surodex pellets had migrated into the inferior AC angle, and in 14 eyes (39%), both Surodex pellets had migrated into the AC. In contrast, no migration was observed with AC placement. It should be noted that in eyes with PC placement, Surodex was placed preferentially within the ciliary sulcus, as opposed to in-thebag placement, to reduce the risk of pellet loculation within the capsular bag, and this may have resulted in a higher instability rate. In addition, deep placement into the depths of the ciliary sulcus was not achieved in most cases because of poor visibility and access behind the iris diaphragm. Although pellet migration occurs with PC placement, the overall efficacy of Surodex does not appear to depend on placement site, because this study was unable to show any significant differences in aqueous flare reduction between AC and PC placement. This suggests that exact localization of Surodex within the anterior segment may not be necessary to achieve adequate antiinflammatory efficacy. This
Tan et al 䡠 Comparison of Placement of Two Surodex after Cataract Surgery has importance in instances of poor anterior segment visualization. In addition, pellet migration within the visual axis does not seem to cause visual symptoms, as evidenced by the two group C patients in whom Surodex had migrated into the visual axis, between the IOL and the posterior capsule. Neither patient reported visual degradation on direct questioning. This study also reinforces previous safety findings of the first trial regarding ocular insertion of Surodex. Placement of two Surodex pellets either in the AC or in the ciliary sulcus of the PC did not result in significant intraocular complications. In particular, only one eye within the Surodex arms had an initial transient IOP rise, whereas minimal irritative or local effects related to Surodex insertion were seen. Despite the fact that in almost one third of cases residual pellet remnants remained present at 1 year, no late IOP rise was encountered. In addition, migration of Surodex into the visual axis, which occurred in two group C eyes, did not result in any visual disturbance (Fig 5). All trial patients attained 6/12 vision at 3 months, with the exception of one patient with early posterior capsule opacification. Posterior capsule rupture with vitreous loss and necessitating AC IOL placement also does not appear to be a contraindication for Surodex insertion. Corneal endothelial cell counts suggest that two Surodex inserted in the AC do not result in significant endothelial cell damage. In addition, the finding that no significant differences in mean inferior endothelial cells counts were encountered between group A and group B patients suggests that insertion of two Surodex pellets in the inferior angle of the AC does not lead to adjacent inferior endothelial cell damage or loss. These results are similar to the previous clinical trial with one Surodex insertion. In summary, this study demonstrates that placement of two Oculex Drug Delivery System devices containing dexamethasone in the eye after intraocular surgery is safe. Two Surodex pellets are equally effective, but not more effective than one Surodex, in reducing aqueous flare after cataract surgery as measured by laser flare photometry, and although anterior migration occurs with PC placement of the DDS, no difference in efficacy occurs. Surodex is currently approved for clinical use in Singapore, China, and Mexico, and phase III Food and Drug Administration clinical trials have been completed in the United States. Clinical trials on an antibiotic DDS are currently underway, and it is anticipated that the reality of cataract surgery without the need for eyedrops is close at hand. The use of Oculex DDS technology for controlled drug delivery in many other forms of ocular surgery and ocular diseases is also being explored.
Acknowledgment: The authors thank Professor Ng Tze Pin, Department of Community, Occupational and Family Medicine, National University of Singapore, for statistical consultation.
References 1. Munton CGF, Phillips CI, Martin B, et al. Vicryl (Polyglactin 910). A new synthetic absorbable suture in ophthalmic surgery. Br J Ophthalmol 1974;58:941–7. 2. Tan DTH, Chee SP, Lim L, Lim ASM. Randomized clinical trial of a new dexamethasone delivery system (Surodex) for treatment of post-cataract surgery inflammation. Ophthalmology 1999;106:2:223–31. 3. Kraff MC, Martin RG, Neumann AC, Weinstein AJ. Efficacy of diclofenac sodium ophthalmic solution versus placebo in reducing inflammation following cataract extraction and posterior chamber lens implantation. J Cataract Refract Surg 1994;20:138 – 44. 4. Mishima S, Tanishima T, Masuda K. Pathophysiology and pharmacology of intraocular surgery. Aust N Z J Ophthalmol 1985;13:147–58. 5. Hogan MJ, Kimura SJ, Thygeson P. Signs and symptoms of uveitis. I. Anterior uveitis. Am J Ophthalmol 1959;47(5 pt 1):155–76. 6. El-Maghraby A, Marzouki A, Matheen TM, et al. Reproducibility and validity of laser flare/cell meter measurements of intraocular inflammation. J Cataract Refract Surg 1993;19: 52–5. 7. Sawa M, Tsurimaki Y, Tsuru T, Shimizu H. New quantitative method to determine protein concentration and cell number in aqueous in vivo. Jpn J Ophthalmol 1988;32:132– 42. 8. Martin RG, Sanders DR, Van Der Karr MA, DeLuca M. Effect of small incision intraocular lens surgery on postoperative inflammation and astigmatism. A study of the AMO SI-18NB small incision lens. J Cataract Refract Surg 1992; 18:51–7. 9. Altamirano D, Mermoud A, Pittet N, et al. Aqueous humor analysis after Nd:YAG laser capsulotomy with the laser flarecell meter. J Cataract Refract Surg 1992;18:554 – 8. 10. Alio´ JL, de la Hoz F, Ismail MM. Subclinical inflammatory reaction induced by phakic anterior chamber lenses for the correction of high myopia. Ocular Immunol Inflamm 1993;1: 219 –23. 11. Vita RCA, Campos M, Belfort R Jr, Paiva ER. Alterations in blood–aqueous barrier after corneal refractive surgery. Cornea 1998;17:158 – 62. 12. Ku¨ chle M, Nguyen NX, Naumann GOH. Aqueous flare following penetrating keratoplasty and in corneal graft rejection. Arch Ophthalmol 1994;112:354 – 8. 13. Ku¨ chle M, Nguyen NX, Seitz B, et al. Blood–aqueous barrier after mechanical or nonmechanical excimer laser trephination in penetrating keratoplasty. Am J Ophthalmol 1998;125:177– 81.
2181